Discovering the diagnostic biomarkers underlying Type 1 diabetes and Celiac disease by integrating transcriptomics and machine learning

Abstract

Abstract An immune-mediated disease with a long latency period is type 1 diabetes (T1D). The beta cells in the pancreatic islets die due to antibody-mediated mechanisms in T1D. The most common chronic disorders are celiac disease (CD). A specific serum antibody response characterizes CD, a complicated systemic immune-mediated enteropathy. Due to the immunological driven nature of both CD and T1D, the risk factors are comparable. Both are increasing at a global rate, T1D screening is advised since CD and autoimmune diseases like T1D frequently co-occur. Based on a large number of research, even if the exact pathophysiology is yet unknown. Our understanding of one disease affects the way we treat as we discover about another, and vice versa. The goal of this study is to look at the connections between both conditions to identify biomarkers that could potentially utilized to diagnose the two. We conducted a comprehensive analysis of the Differentially Expressed Genes (DEGs) identified in the samples using a range of bioinformatics techniques and machine learning algorithms, built a network for the biological interactions, and discovered 3 potential diagnostic genes with statistical values (NAA15, RPL21, and HCLS1) as legitimate candidate genes as the biomarker for the diagnosis of T1D and CD.